MCP Fundamentals 🧱

Intro

👨‍💼 Hello, my name is Peter the Product Manager. I'm here to help you get oriented and to give you your assignments for the workshop!
Welcome to MCP Fundamentals! In this workshop, you'll dive into the world of the Model Context Protocol (MCP)—an open protocol designed to seamlessly connect AI models with external data sources and tools. Think of MCP as the "USB-C port" for AI applications: it standardizes how LLMs (Large Language Models) access context, resources, and capabilities from a wide variety of sources.

Internet

Your Computer

MCP Protocol

MCP Protocol

MCP Protocol

Host with MCP Client
(Claude, IDEs, Tools)

MCP Server A

MCP Server B

Local
Data Source A

MCP Server C

Remote
Service B

Remote
Service C

Throughout this workshop, you'll learn:
  • The core architecture of MCP: hosts, clients, and servers (see architecture)
  • How JSON-RPC powers communication and stateful sessions between components
  • How to expose resources, prompts, and tools to LLMs using MCP
We'll focus on the fundamentals—no authentication or authorization required for our exercises—so you can get hands-on experience with the protocol's most important features.
We'll be using the MCP TypeScript SDK to build our MCP server and the MCP Inspector to test it.
If you're curious about the details, I highly recommend checking out the official MCP specification and the introduction. These resources provide a deep dive into the protocol, its design principles, and practical implementation tips.
Ready to connect your AI to the world? Let's get started!
Edit on GitHub